top of page
indigoChart Menu
Data Migration
Consolidating and landing data from legacy databases into the intended modern data platform is a standard procedure in data modernization projects. This may be a continuous process or a one-time move. Regardless of the scenario, we can help with the migration of data and metadata from legacy source databases like SQL Server, Oracle, Teradata, Netezza, Postgres, and MYSQL to Snowflake and Databricks
Here’s our Actionable Data Migration Strategy that outline factors to be considered for your data and metadata migration strategy with Snowflake as the chosen modern data platform and using our data migration accelerator iC-DataElevate.
Step 1. Analyze the current environment attributed by
1. Data Volume
2. Data Pattern or velocity
3. Data Type or format (Structured, Semi-structured, Unstructured)
4. User classification
5. Critical workloads (Data Engineering/Data Science or AIML / Data Visualizations)
6. Data Object and conversions for tables, functions, views, stored procs
Step 2. Create Migration plan that takes into consideration
1.Prioritized Databases/schemas
2. Prioritized Applications
3. Critical workloads
4. Smaller sized databases or those down the priority
5. As-is, Re-engineered, or Hybrid approach
6. Leverage data migration accelerators such as indigoChart’s iC-DataElevate to reduce migration times
Step 3. Plan for workloads supported by the chosen data platform along
1. Solution Architecture
2. Databases definitions
3. Compute definitions
4. Role Based Access Control (RBAC) strategy
5. Data Sharing and Replication strategy
6. Data Security Architecture
7. Data Extraction Approach
8. Data Transformation approach
9. Semantic layers and BI tools
10. AI and Gen AI Architectures
Step 4. One time data extraction, loading, and validation
1. iC-DataElevate for Data Extraction
2. Architecture and planning
3. Data Objects
4. Coding (views/stored procs/functions) - migration approach
5. iC-DataElevate for Data Loading from source to target
6. Data loads comparison, Row counts, Hash counts
7. Data quality and distinct value checks, random value checks
Step 5. Performance Checks and cutover planning
1. Query runs and comparisons on legacy vs new environments
2. Performance checks
3. Output checks
4. Compute sizes and setups per requirements
5. Cutover timelines planning
6. Redirect data to new environments
Step 6. Post migration validations
1. Daily loads check on new environments
2. Compute performance, Load, and Transformation check
3. Peak data reads performance checks
4. New use cases checks and support
5. Datawarehouse Validations on Legacy Primary against New Target
6. Datawarehouse Validations on New Primary against Legacy Source
7. Legacy Datawarehouse Sunset process
bottom of page